Symbotic Inc. (NASDAQ:SYM) Q3 2025 Earnings Call Transcript

Symbotic Inc. (NASDAQ:SYM) Q3 2025 Earnings Call Transcript August 7, 2025

Operator: Thank you for standing by. My name is Gil, and I will be your operator for today. At this time, I would like to welcome each and every one of you to the Symbotic Third Quarter 2025 Financial Results. [Operator Instructions] It is now my pleasure to turn today’s call over to Mr. Charlie Anderson, Vice President of Investor Relations. Please go ahead.

Charlie Anderson: Thank you. Hello. Welcome to Symbotic’s Third Quarter of Fiscal Year 2025 Financial Results Webcast. I’m Charlie Anderson, Symbotic’s Vice President, Investor Relations. Some of the statements that we make today regarding our business operations and financial performance may be considered forward-looking. Such statements are based on current expectations and assumptions that are subject to a number of risks and uncertainties. Actual results could differ materially. Please refer to our Form 10-K, including the risk factors. We undertake no obligation to update any forward-looking statement. In addition, during this call, we will present both GAAP and non-GAAP financial measures. A reconciliation of GAAP to non-GAAP measures is included in today’s earnings press release, which is distributed and available to the public through our Investor Relations website located at ir.symbotic.com.

On today’s call, we’re joined by Rick Cohen, Symbotic’s Founder, Chairman and Chief Executive Officer; and Carol Hibbard, Symbotic’s Chief Financial Officer; as well as Izzy Martins, our CFO successor designate, who will become CFO effective August 9th. These executives will discuss our third quarter fiscal year 2025 results and our outlook, followed by Q&A. With that, I’ll turn it over to Rick to begin. Rick?

Richard B. Cohen: Thank you, Charlie. Good afternoon, and thank you for joining us to review our most recent results. In the third quarter, we once again delivered strong financial results, while driving key operational progress. Revenue increased by 26% year-over-year, and importantly, we maintained improved margins, thanks to disciplined cost control and solid project execution. I want to commend our team for their continued focus and high-quality delivery. Our automation systems are also showing tangible results for our customers. Adoption continues to scale, and we recently set a record processing over 6.5 million cases in a single day through our operational systems. This level of throughput highlights the real value we’re delivering to our customers.

On the development side, we also made solid progress on our integration of Advanced Systems and Robotics or ASR business. Notably, we’ve now identified sites for the development of the first prototypes of the next-generation solution. Installation is expected to begin early next calendar year, keeping us on track with our road map. From a technology standpoint, we’re driving innovation across the stack. A key example of this is our teleoperations capability, which enables remote operators to use our bots to reposition misaligned cases that have shifted out of place during inbound processing. Before this capability, these tasks sometimes required manual intervention during scheduled downtime. Recently, we reached an important milestone, our first operational day at a large site with 0 manual repositioning.

Using machine learning, we are training our bots to automatically replicate these tasks, minimizing both downtime and labor needs. Our goal remains clear, smarter bots equipped with cameras, LiDAR and advanced GPUs, enabling even greater efficiency and value. Finally, on the innovation front, yesterday, we announced a major product milestone, the debut of our next-generation storage structure. This marks what I believe is one of the most significant product upgrades in our company’s history. This new structure substantially increases our already exceptional storage density, allowing customers to store considerably more products in the same space and/or reduce their overall storage footprint. A more compact structure also speeds case handling as bots travel shorter distances.

Equally important, the new structure introduces preassembled precision manufactured subcomponents, reducing on-site assembly parts by over 90%. It also features a unique leveling system that minimizes floor preparation. Together, we believe these improvements will accelerate deployment and enhance scalability, thus making the next-generation structure a game changer for our business. Additionally, we plan to apply this structure across all aspects of the supply chain from distribution centers to perishable goods environments to e-commerce and micro fulfillment hubs. Customer response to next-generation storage has been very positive. In fact, we began signing projects that include the new storage structure in our fiscal third quarter. We believe this next-gen storage structure supports our path to unlocking even higher margins and long-term value creation.

In summary, Symbotic remains well positioned. Our strong and growing product portfolio spans multiple levels of the supply chain supported by a $22.4 billion backlog and a healthy balance sheet. Thank you to our team for their outstanding work this quarter and to our customers and shareholders for their continued support. I will now turn it over to Carol, who will discuss the quarterly financials before I come back to introduce Izzy Martins, who’s taken over as CFO next week and will provide the forecast. Carol?

Carol J. Hibbard: Thanks, Rick. Before I begin, I want to thank the entire team at Symbotic. It has been a pleasure to serve as CFO with you. Symbotic is in a strong position, and I look forward to supporting the team during the transition period. Now to the results. Third quarter revenue grew 26% year-over-year to $592 million, with revenue growth driven by solid progress across our 46 systems in deployment, expansion of the number of systems in operation and continued progress on our ASR development. Our net loss for the third quarter was $32 million versus a loss of $27 million in the third quarter of fiscal year 2024. Adjusted EBITDA in the third quarter of $45 million was well above our forecast and up significantly year-over-year from $3 million in the third quarter of fiscal year 2024.

In terms of the backlog, our backlog of $22.4 billion remained in a strong position. The sequential decrease from $22.7 billion last quarter was due to revenue recognized in the quarter, partially offset by final pricing on projects started. In our third quarter, we began 5 new system deployments. We also had 5 systems go operational in the quarter, bringing our total to 42 operational systems. As we mentioned last quarter, we view the most critical portion of deployment as the time between the start of installation and when the system goes operational at acceptance. This is the time in which we see the most revenue and profit contribution. In the third quarter, we again saw improvements against this metric. During the third quarter, we had a Phase 1 completion for our largest customer that was nearly 2x the size of our historical average for Phase 1 deployments for this customer.

Despite that added size, the time between start of installation and going operational was only slightly higher than our historical average and significantly less if we normalize for size. We also completed 2 Phase 2 deployments for our largest customer that both took over 20% less time between start of installation and acceptance than the historical average of their predecessors. With the continued growth in operational systems, we saw our software revenue more than double year-over-year to $8.1 million, and operations services revenue grew 54% year-over-year to $24.9 million. Turning to margins, system gross margin continued last quarter’s trend of significant year-over-year improvement, thanks to strong cost control and project execution.

Gross margin on software maintenance and support exceeded 75%, as it continued its trend towards software industry margins as we gained scale. And in Operations Services, we came in with a slight profit. Operating expenses on a GAAP basis were up sequentially, primarily due to a $16 million restructuring charge from the workforce reduction associated with our acquisition of the Walmart ASR business. On a non-GAAP basis, operating expenses were down sequentially, coincident with activity from the development phase of ASR. We finished the quarter with cash and equivalents of $778 million, which decreased from $955 million in the second quarter, primarily due to the timing of cash receipts. In summary, we had a strong financial quarter matched with continued operational progress.

I’m now going to hand it back to Rick, who will introduce Izzy. Rick?

Richard B. Cohen: Thank you, Carol. On behalf of everyone at Symbotic, I want to thank you for your contributions, namely your leadership in driving improved project execution, while controlling costs, which we are reaping the benefits of today. We are fortunate to have a good amount of overlap between Carol and Izzy, who joins us from Avis Budget Group, where she served as CFO and previously served as Executive Vice President, Americas, where she oversaw a market segment with over $9 billion of revenue. Izzy brings to us strong track record of financial leadership and operational expertise, which is critical for the next phase of the company. We’re thrilled to have her here. With that, I’ll turn it over to Izzy, who will introduce herself and provide the forecast. Izzy?

Izilda Martins: Thank you, Rick. I’m honored to join Symbotic, a recognized leader in supply chain transformation through robotics and innovation. It’s exciting to be a part of an organization that consistently pushes the boundaries of operational excellence. I also want to extend my sincere thanks to Carol for the warm onboarding experience and for generously sharing her deep institutional knowledge over the past month. Her guidance has been invaluable, and I am grateful for her continued support during this transition. I look forward to partnering with our exceptionally talented leadership team, as we enter our next phase of growth with a strong focus on delivering sustained value for our customers and our shareholders. Finally, I am eager to engage with many of you in the investment community in the coming months.

I welcome the opportunity to share our vision for the future. Now turning to the forecast. As Rick highlighted earlier, we are in the process of bringing to market a next-generation storage structure that we fully expect will set the new standard for our customers. Over the long term, we anticipate this innovation will unlock meaningful opportunities, enabling more efficient deployments and supporting a higher margin profile for Symbotic. In the near term, we anticipate an adjustment in the timing of several previously planned deployments as efforts are realigned to support the transition to the next-generation storage structure. As a result, the quarter-over-quarter sequential growth we’ve seen in recent periods will be less pronounced in the fourth quarter.

Looking ahead to the first half of 2026, we believe we could see similar impact as we adjust deployment schedules to accommodate a transition to the new structure. It is important to note that this does not affect our overall backlog. In fact, we believe the new structure rapidly — rapid assembly characteristics will allow us to scale faster over time. With that in mind, for the fourth quarter of fiscal 2025, we expect revenue between $590 million and $610 million and adjusted EBITDA between $45 million and $49 million. With that, we now welcome your questions. Operator, please begin the Q&A.

Q&A Session

Operator: [Operator Instructions] Today, our first question comes from the line of Andy Kaplowitz with Citi.

Andrew Alec Kaplowitz: So maybe you can help us quantify this next-generation storage technology in terms of how first, how much faster can I make installation times? And how do you think about the retrofit opportunity set for Symbotic of this technology? Would you expect, for instance, Walmart to do a bunch of retrofitting of systems that you’ve already completed with them? And then will they actually pay you for that?

Richard B. Cohen: Yes. So I’ll take that. So no, we don’t expect Walmart to do retrofits. The beauty of this system is that it can run side-by-side with the old structure, but it takes less room. So in truth — some of the projects we’re working on with Walmart will actually be smaller because of the new structure, but actually allow them to sell us more product in the same building or additional capabilities. For instance, we could put in the same building that was going to just do replenishment to stores. We could put a break pack system in and we could actually put an ASR system in that might actually begin to start to look a little bit like an e-commerce solution. So the capabilities are hugely expand our business. And that was one of the things that we wanted to talk about that smaller — the history of all innovation is always comes down to whether it’s transistors or many other things, miniaturization is, for the most part, always a huge positive.

And that’s really what we’re doing. Likewise — so they will continue to use the old structures and the buildings that they have. But like some of the Phase III buildings that we’re doing, we actually have a new structure right alongside the old structure. And that took some real development work on our part, but we didn’t want the customers to have to go backwards. So they’re very excited about the positives.

Carol J. Hibbard: And Andy, I’ll follow up on your question related to deployment time lines. So as you know, we follow a key metric, which is install to actual deployment time line. historical, that’s been about 12 months, and we’ve seen step-downs in that on our existing core. Given the reduction in parts and the streamlined assembly that we plan in terms of that build, you would likely see that time line begin to reduce, and we’ll share that progress as we roll out the design and begin deployment.

Andrew Alec Kaplowitz: Rick, I just have one follow-up. Like so the way you think about the growth like line, it’s less steep for the next few quarters and then it can be even steeper than it was past that. That’s the way to think about it in a positive direction, correct?

Richard B. Cohen: Yes. That’s exactly the way we thought about it. And when we started looking at the development of this, we thought about that. But for instance, we’ve now had a number of inquiries about perishables because the cost of construction — but the facility itself is so expensive today to be able to retrofit a perishable facility, this may turn out to be the best retrofittable perishable facility that exists. It’s a brand-new product for us.

Andrew Alec Kaplowitz: And then, Rick, I just wanted to ask you about — you’ve increased the sales force. You talked about that last quarter. Backlog is still in this sort of $22 billion to $23 billion range. And I think you talked about still expecting to get some new customers or a new customer this year. Is that still the case? And is it more likely to come at Symbotic or GreenBox?

Richard B. Cohen: I think it’s going to come at both. We really — we announced the new structure yesterday, but we’ve been marketing the new structure probably for the last 3 to 6 months with customers telling them it might be coming. And then we’ve started doing some designs. The first installation will actually be at in Bajio in Mexico for a big Walmart project, and then they’re going to start to come after that. But we’ve just really started marketing that in the last 3 to 6 months. And so, our inbound discussions with new customers is accelerating as a result of it.

Operator: Your next question comes from the line of Joe Giordano with TD Cowen.

Joseph Craig Giordano: Maybe I’ll start just on that structure. I’m sure we all watch the videos. I get it’s clear why this would be faster to put together. I’m just curious, maybe you could talk me through like the main reasons why it’s more dense. I think that’s a little bit harder to just tell from the video, like why it’s necessarily smaller. And can you help us like what does this mean for like the price of these things?

Richard B. Cohen: So the reason that it’s smaller is that it’s a cantilever structure and believe it or not, the posts that supported the original structure took up 10% of the — we think of it as linear storage space. And then the other thing you’ll see in the video is because we have fingers or times, we can actually put 3.2 shells per level versus 2.2 [indiscernible] per level, which is a 30% increase. And that’s why the 40% is a pretty good number for storage density. So what we basically did is took all the air out of the structure and increase the cubic density and the linear density of the space, which outside of me is probably the most boring thing in the world for most people to talk about, but it has huge impacts to the customer, and we’ve actually patented it.

It’s probably the first new structure everybody else is using traditional warehouse rack more or less to do their automation. And so this is a brand-new structure, which we have lots of patents on, and it goes up very quickly. So we have a bunch of new design structural engineer people that have helped us with this. So that’s the first answer to your question. And then we designed it so it could go up in modular pieces. There’s no — the other system had shells, we would run about 1 million rivets for structure, and there’s no rivets in the structure. So it’s a big, big change, much faster than [ install ].

Joseph Craig Giordano: And the implications on cost for something like that?

Richard B. Cohen: So the implication on cost is we’re value pricing this. So for the customer, we think we can save them more money because they don’t need as much space. And for us, we expect to make higher margins.

Joseph Craig Giordano: Higher margins and maybe lower dollars, but maybe you can protect the dollar — the gross profit dollars. Is that the right way to think about it?

Richard B. Cohen: Maybe not lower — maybe not lower dollars.

Carol J. Hibbard: And I think another way to think about it, Joe, is it’s going to vary by customer because we’re building this in less space. The customer will have a decision to make of they might want to maximize that additional density and utilize that because we’ve created them capacity. So I think it will vary as they look at how they want to go and deploy.

Joseph Craig Giordano: And then just mechanically in the model here, like how should we think about stock-based comp going forward? Because it’s — I think it was initially expected to trend lower. It’s been pretty high the last couple of quarters, and it’s like a huge add back to adjusted margins. So where do you see the glide path on that?

Carol J. Hibbard: Yes. I’d say the glide path of what you saw last quarter and where we are this quarter, we will continue that trend at least for the next few quarters going forward. Really, the increase has been a focus on, obviously, some of the talent you’ve seen us bring in and then our headcount has gone up. That also is — some of the increase was related to just our overall acquisition associated with ASR in there as well.

Operator: Your next question comes from the line of Nicole DeBlase with Deutsche Bank.

Nicole Sheree DeBlase: Maybe just starting with the big step-up that you saw in software gross margins, pretty impressive this quarter to be north of 70%. Any comments on the drivers of that? And should we now view that above 70% level as sustainable moving forward?

Carol J. Hibbard: Yes. Thanks for the question, Nicole. I’d say given that this is the second or third quarter in a row, where we’ve seen the step-up, I would assume that 70% level continues going forward. So this quarter, we added 7 new acceptances into the quarter. And so we’re really seeing the benefit of scale. Last quarter, we added 8 systems that hit complete. And this quarter, we saw 5. And so we’re really seeing the benefit of being able to scale across the software team.

Nicole Sheree DeBlase: Got it. That’s great. And then secondly, on free cash flow, I think it was a use of cash this quarter, which surprised me a little bit. Can you just provide some more color on that? Maybe it has something to do with the new structure, et cetera? And any thoughts on free cash in 4Q?

Carol J. Hibbard: Yes. The free cash flow was not associated with the new structure. It was entirely timing of receipts. So we signed projects later in the quarter, and we’ll see that step back up in the fourth quarter. One thing to note on free cash flow going forward, which is tied to the new structure. And so you’ve seen our CapEx relatively flat around $12 million to $15 million a quarter. You’re going to see a step up over the next couple of quarters as we make the investments to support the build of the new design. So all of the times that Rick talked about, we’re investing in the equipment to produce those. So you’ll see a step-up in CapEx over the next couple of quarters.

Nicole Sheree DeBlase: Okay. And any way to quantify the step-up in that CapEx approximately?

Carol J. Hibbard: Yes. I’d say you’ll see CapEx about 2x what you saw this quarter.

Operator: Your next question comes from the line of Mark Delaney with Goldman Sachs.

Mark Trevor Delaney: First, on the new storage structure, thanks for all the details you shared already on that topic. I’m hoping you could be a little clearer when do you expect to start providing this to customers? You mentioned Mexico being the first location with Walmart, but I wasn’t quite sure when you expect that to begin. And if you could also help us understand of the backlog, is your expectation that most of the systems you haven’t started yet are going to be with this new structure? Or do you need to go out and negotiate that?

Carol J. Hibbard: So I’ll go ahead and start. So yes, as Rick indicated, we already have customers who have signed. In fact, all 5 of our starts for this quarter, we signed those assuming the next-gen structure configuration. And so our expectation is our customers going forward will all be associated with the next-gen structure. We expect the first one to begin installation mid next fiscal year, so our mid fiscal year 2026.

Mark Trevor Delaney: Okay. And then the sort of that sequential moderation you spoke about, you’re still starting some — are there still going to be some system starts just with the old structure until the new one is ready? Or are new starts going to be effectively near 0 until middle of next year?

Carol J. Hibbard: Yes. No, new starts won’t go to 0. So that’s a good clarification point. So in fact, we had 5 new starts this quarter. All 5 of them are going to proceed with the new structure. And so with the design activity we spent over the last several quarters, we’ll go ahead and incorporate the CapEx and be able to begin installation of those sites in mid-2026. You are not going to see a stop of new South signatures as we go forward based on the new design. You saw a slowing of them this quarter as we had customers waiting for some of the deployments that they would have started and that’s what you’re seeing impactful in our guidance for the fourth quarter and what we referenced for the first half of ’26.

Mark Trevor Delaney: Okay. Just one other from me, if I could, on GreenBox. When Symbotic announced the GreenBox deal 2 years ago, the plan, I believe, was to start installation of all of those systems within 6 years. I think it’s been a relatively limited number of systems that Symbotic has started for GreenBox thus far. So I’m hoping to better understand what your expectation is at this point? And do you still expect to begin installation of all of those GreenBox systems within the prior 6-year target?

Carol J. Hibbard: And so as you indicated, we’ve got 3 systems currently in deployment. And you actually saw within the quarter, we’re seeing installation ramp up. And so you’re seeing revenue from GreenBox start increasing on those systems. So we have indicated over the last couple of quarters, we saw slower-than-planned starts of GreenBox systems. We expect to see now that we’ve got the CEO in place and the sales team in place, we expect to start seeing that move forward in the coming quarters.

Operator: Your next question comes from the line of Colin Rusch with Oppenheimer.

Colin William Rusch: There’s been an awful lot of innovation in and around the energy storage space. I’m just curious about how much leverage and how much opportunity there is to optimize bot performance uptime some of the design as you start to see some of those chemistries mature at a level that you get comfortable with.

Richard B. Cohen: Yes. So the new structure being denser will mean that the bot trips will be shorter. And so we are expecting to see somewhere a pretty substantial increase in bot transactions per minute, per hour. So if we have less travel time, we’ll need less bots, and that will reduce the cost of a system. The second thing is that we’re also — while we’re talking about bots, putting LiDAR and upgrading the vision chips and the ability for the bots to compute what they’re looking at. So all of those things are — there’s a big push on innovation, which will happen, I think, in — certainly within the next year, where bots are going to be able to do substantially more work than they do right now.

Colin William Rusch: Okay. I’ll follow up afterwards on the battery impact. And then I guess from a dexterity and kind of material handling perspective, obviously, you guys are the clear leader in terms of the number of materials and the types of materials that you guys can handle. I guess, are there areas that you guys are trying to innovate on right now that might open up incremental opportunities? And can you give us a sense of timing in terms of just some of the materials that are difficult to manage that you guys might be able to unlock here over the near term?

Richard B. Cohen: When you say materials, one of the — I’m not sure exactly what you mean. One of the clear targets that we’re going after is perishables, because a lot of the — because of $500 a square foot for some of these buildings, smaller buildings are hugely valuable to the customer and they get the savings on the building and we get to be able to sell them a very valuable system. So that’s one thing. Same thing on the freezer. Freezers are not that hard for us to do, but the storage density has made people look at some of the prior providers and say, oh, maybe I can expand my existing freezer without having to add as much space as I thought. So where we’re really going after value creation is ambient was all about the ability for us to handle so many more items than traditional systems.

But on the perishable side, there’s also the value they’ve created just because the facility itself are so expensive even before they put our system in. So that’s one area that we’re definitely going after. We don’t think the temperature has a significant effect on our bots.

Operator: Your next question comes from the line of Derek Soderberg with Cantor Fitzgerald.

Derek John Soderberg: Rick, just a question on the competitive environment. I think we’ve all seen videos online. There’s humanoid robots out there. I think one of your competitors has kind of robotic unloading technology from a semi truck. Any sort of change in the competitive environment? I know there was a question earlier on some of the technology, but it seems like the pace of innovation has really stepped up. Any new concerns or any new technologies out there?

Richard B. Cohen: I mean we’re always concerned, but the response that we’re getting — so the truck unloading is really a very specific thing, and we’re in touch with a lot of those folks. It might be an add-on to our system. That would actually make our system more valuable. The humanoid robots are very good at manufacturing, but our robots are going 25 miles an hour. So the humanoid robots are going to be very good stationary or slow-moving machines. They’re — and basically, what they incorporate is dexterity and fingers and good vision. We do that with our bots, but because we’re picking up cases, we’re less concerned with the humanoid robots. So I think that’s a different aspect of automation that we’re going after. We’re seeing more and more requests for us to look at things that people had previously used our competitors for.

So I would say we feel good about where we are and we think our TAM and our market share is going to grow. But until it does, I can’t prove it out, but that’s what I think.

Derek John Soderberg: Got it. That’s helpful. And then just with the new storage system, I think there were a number of questions on the cost, how it’s going to impact backlog. But just when you combine that with a lot of the changes you guys have made on the EPC front, bringing that in-house, it’s been a while since we’ve gotten kind of an update on maybe the structural gross margins of the hardware piece, the systems piece. Where are we sort of headed with all these changes that you guys have made? What are sort of the structural gross margins in systems? How much more room do we have to kind of continue to see margin expansion from here in systems?

Carol J. Hibbard: Yes. The next-gen structure, Derek, is actually one of the key enablers on our path to 30% system gross margin. So systems-wise, we’re still on that path. And we’ve talked about overall improvements around schedule being a driver, bringing back the EPC was certainly one of the significant drivers to get us where we posted the last couple of quarters. But the next-gen structure, the pure fact that it’s streamlining assembly, reducing the million rivets that Rick mentioned, that’s going to help us on our trajectory towards higher gross margins. And then as we think about the value creation piece of this, as we go out and attract new customers, we’re certainly driving to utilize that to grow the gross margin profile.

Operator: Your next question comes from the line of Greg Palm with Craig-Hallum.

Gregory William Palm: I maybe just want to go back to the quarter because what really, I guess, stood out to us was the operating leverage. The incremental margins were extremely high in the quarter. So I’m just curious if there’s anything sort of one-time, any benefit in the quarter that you saw? And just as we kind of think about fiscal ’26, how should we kind of think about that going forward?

Carol J. Hibbard: Yes. From a gross margin perspective, the step-up in software was certainly a good contributor. And when I think about mix, so we had a higher contribution this quarter from our advanced systems and robotics portion of the business, higher margin content than some of our historical contracts have been. So you’re seeing that. And then the other lever, which continued to be beneficial this quarter is that time line to deployment that we indicated from the start of install to final acceptance, we’re seeing continued improvements on how long it’s taking us to build these. And so you’re seeing that reflected in our gross margin.

Gregory William Palm: Okay. That’s helpful. And then I guess just maybe one clarification. In terms of the new storage system, I’m curious, has this been a part of the product road map for a number of years? Was this something that was maybe more directed by one of your customers? I’m just kind of curious to sort of know the time line or the history and maybe some of the background.

Richard B. Cohen: Yes. So I’ve been working on this for a couple of years. And I have — I did my first drawing of it. But the way I describe it, I feel like Frank Gehry, I actually had a design and I needed some engineers that could actually build it. And so this has been in the works for 2 years. We’ve teased customers with it, tested them, asked them if they wanted it. They got very excited. And so really, it’s been the last year that we — I mean, if you come visit us, we have the test systems right down the street now and it’s just — it’s a magnificent new product.

Gregory William Palm: Okay. Carol, best of luck going forward and Izzy, looking forward to working with you.

Operator: Your next question comes from the line of Ken Newman with KeyBanc Capital Markets.

Ken Newman: Carol, maybe a little bit of help and piggybacking off the margin question just now. Maybe can you help us think about systems gross margin sequentially from the third to the fourth quarter? And just given all the shift in the production schedules, can that still stay flat? Or would you expect that to still be up sequentially 3Q to 4Q?

Carol J. Hibbard: Yes. Our expectation is we’ll see systems gross margin about where they were in the third quarter. We’ll see that continued trajectory. And as we mentioned on software earlier, we believe the 70% level on software gross margin is where we’re going to head.

Ken Newman: Got it. Okay. And then maybe one more on the cash flow. I appreciate the help on the CapEx and some of the moving pieces on timing of receipts. You do have a pretty high portion of your revenue going towards R&D. I’m just curious if there’s anything you can kind of parse out on what’s potentially able to be capitalized going forward with this new tax bill and how to think about structural free cash flow through the cycle with the new policy.

Carol J. Hibbard: Yes. I’d say for R&D, the levels that you’re seeing today will continue. We actually got the benefit over the last couple of quarters of a substantial portion of R&D is now actually moving to our contracted associated with the advanced systems and robotics. So you’re seeing that as a lever, too, in our R&D as what you saw in this quarter. So R&D from 3Q to — from 2Q to 3Q, you saw a slight reduction there. And that’s because we’re able to charge that off to our contracted R&D value.

Operator: [Operator Instructions] Your next question comes from the line of Mike Latimore with Northland Capital Markets.

Michael James Latimore: Okay. Great. In terms of just new starts, should they be kind of in this mid-single-digit range for a while until you get rolling with the next-gen storage system? Or how should we think about just kind of new start trajectory?

Carol J. Hibbard: I’d say this is the quarter where we saw a fewer number of starts as we were waiting to make sure we got the design finalized. And so you’ll see new starts step up. We’ve talked about the average of mid- to high single-digit new starts. Obviously, last quarter, we hit a record high of 10 new starts. This quarter dropped down to 5. But you should see that step up again going forward now that we have the design commercially available, and that’s what a couple of our customers were waiting on in terms of moving forward with deployment.

Michael James Latimore: Great. And then just a clarification on the guidance. When you talk about the first half of ’26, is the implication that it should be a similar quarterly revenue level to what you’re seeing in the — we’re guiding to in the September quarter?

Carol J. Hibbard: Yes. As you know, we don’t guide beyond the next quarter. But if you look at what we put in our Q for the next 12 months associated with RPO consumption, similar level to what we’re guiding for 4Q would be where we — that would put us for the first half.

Operator: Next question comes from the line of Matt Summerville with D.A. Davidson.

Matt J. Summerville: Just a couple of things real quick. Does this new storage system, does that help accelerate new customer acquisition for you guys? Will you look to maybe change that cadence or that targeting of 1 to 2 a year to something greater than that? And can you just give a little bit more of a detailed update on GreenBox? Is there anything you’re able to disclose on tenants? And then I have a follow-up.

Richard B. Cohen: Yes. I mean on new customers, we expect that to grow faster than it is right now. We’re — the customers we’re in dialogue with is at the highest level it’s ever been. It gives us the ability — new structure gives us the ability to sell smaller systems and bigger systems, both at the same time and it gives us a lot more flexibility. So we would expect that to accelerate. On GreenBox, we’re continuing to roll out sites and adding on to the organization there to rapidly grow the sales force. We’re in discussions with a lot of folks. Some folks are looking at Symbotic systems and some folks are actually looking at GreenBox as possibilities. So that actually gives us more variability or more capabilities to lean customers.

But as we start to roll out the GreenBox covering the country, we’re getting significantly more interest as opposed to where you can just do one location versus rolling out a number of locations. And then the first location in Lathrop is scheduled to go live pretty soon. So that will be income producing as well.

Matt J. Summerville: And then with respect to this new innovation that’s been the topic of the discussion today, will you be ramping up any new manufacturing partners for that portion of the system? Or is that not part of this or any new installers relative? I guess I’m trying to understand what is — may change from a manufacturing or installer standpoint, if you will.

Richard B. Cohen: Yes. So we’re working with a couple of new manufacturers to have people that will make basically the tie-ins. The steel is basically — most of that steel is coming from the U.S. It’s not really affected by tariffs. It’s a different design of the steel. And the installation will be our same installation teams. So we’ve actually been using those teams here in Wilmington at the ITC, practicing, installing structures, anybody can come and visit. But the installation is much simpler. We actually the learnings from floor and the learnings from what we’ve been doing for the last couple of years in installation have really gone into once we figured out the new structure, then we actually brought in a bunch of structural engineers and a bunch of construction folks to actually design it to make it much easier to install.

So I would say the way I describe it, what we’re doing here is we went from custom furniture to IKEA. And then I can’t install an IKEA project without my wife’s help. So we still need to train the installers, but we’re pretty — but it’s pretty simple.

Operator: Your next question comes from the line of Rob Mason with Baird.

Robert W. Mason: I had a question on the ASR portion of your business. Carol, I think I inferred from your commentary that the development revenues were up sequentially there. Just I was curious if you could confirm it, maybe give us a feel for what those contributed. And as a follow-on to that, is that revenue at some steady state as we go forward, what you’ll recognize per quarter?

Carol J. Hibbard: Yes. So we were up sequentially from the second quarter to the third quarter. So we talked about last quarter being at mid-single digit. We’re in high single digit in terms of the contribution to the revenue this quarter. And you’ll see that carry on at least in the next quarter or 2. As we indicated in our prepared remarks, we’ve identified 2 of the prototype facilities. So we’ve actually begun ordering equipment. And so as the design progresses, we’re now getting ready to start building out our prototypes. And so you’re going to see that level maintained for the next couple of quarters.

Robert W. Mason: And when the prototype build start, how does that influence the development revenue goes down and it still gets classified in systems, right?

Carol J. Hibbard: That’s right. It’s all classified in systems.

Operator: Thank you, everyone. And that concludes our question-and-answer session for today. I will now turn the call back over to Mr. Charlie Anderson for the closing remarks. Please go ahead.

Charlie Anderson: Yes. Thanks, everybody, for joining our call tonight. We really appreciate your interest in Symbotic and look forward to seeing many of you during the quarter at the various investor conferences we’ll be attending. Thank you, and good night.

Operator: Ladies and gentlemen, that concludes today’s call. Thank you all for joining. You may now disconnect. Have a nice day ahead, everyone.